source: buchla-68k/orig/doc/Utilities/CGREP.MAN@ 0170798

                                CGREP

NAME
        cgrep \- search a file for a piece of context

SYNOPSIS

        cgrep [-mi] [-n -p -/pat] [+n +p +/pat] pattern [ files... ]

DESCRIPTION

        Cgrep takes a chunk of context out of a text file based on a pattern
        that lies inside the context, and rules for specifying the amount of
        context.  If no rules are specified, then cgrep works just like grep
        except that it finds only the first occurrence of "pattern".

        A backward searching rule need not be matched by a forward searching
        rule, and vice versa, however, no more than one rule "per direction"
        should be used.  

        Cgrep does not recognize the grep family metacharacters.  It matches
        exactly what it is given.  The default input is stdin.

        When more than one file is searched, the name of the file in which the
        context is found is enclosed in <> at the top of the text.

OPTIONS

        -n      Begin printing 'n' lines before the first occurrence of
                "pattern". If no 'n' is specified, default is one.

        -p      Begin printing at the beginning of the paragraph 
                containing the first occurrence of "pattern".  (assuming
                that paragraphs are separated by a blank line)

        -/pat   Begin printing at the line containing "pat" nearest 
                before the first occurrence of "pattern".

        +n      Stop printing 'n' lines after the first occurrence of
                "pattern".  If no 'n' is specified, default is one.

        +p      Stop printing at the end of the paragraph containing the
                first occurrence of "pattern".  (i.e. the next blank line)

        +/pat   Stop printing at the first line past "pattern" which
                contains "pat".

        -i      Ignore case.

        -m      Mail mode.  Delete all blank space at the beginning of the
                lines to be printed and insert "> ".  This may be useful
                from within the editor in replying to mail or news.


EXAMPLE

                cgrep -2 +/john harry file1

        searches file1, and prints starting 2 lines before the first
        occurrence of "harry" and ending when it finds a line containing "john"

                cgrep -p +p george file2

        searches file2 and prints the paragraph containing "george"

BUGS

        When using both -m and -i, use separate flags.

        Results are unpredictable if multiple rules for searching the same
        direction are given, though it will probably use the latter.

        If the pattern is found in the first of multiple files,
        the filename is not printed at the top.

        If a backward pattern search is done on a non-existent pattern,
        the backward search is ignored.

        Line length is limited to 256 characters.

        Send all comments, etc to iannucci@sjuvax.UUCP